I read about quasi-experimental design and the variations it has such as pre-test post-test non-equivalent groups, and also experimental designs. Having compared these two, I ascertained that the difference between this variation of quasi-experimental design(pre-test post-test non-equivalent groups) and experimental design is randomization. However, what I couldn't understand was:

Why should quasi-experimental design even exist when we can do randomization easily and the randomization has more benefits than matching the participants?

Honestly, I search for the answer but I couldn't find any, and obviously that is why I'm asking a question, yet if you find any alike question I would be happy if you refer me to it. I genuinely appreciate it if you refer me to at least one resource as well. The reference book for the discrimination between these two designs: Introduction to Research Methods and Data Analysis in Psychology, Third Edition by Darren Langdridge and Gareth Hagger-Johnson


1 Answer 1


The first paper I found searching for "advantages of quasi-experimental design" was this one:

Schweizer, M. L., Braun, B. I., & Milstone, A. M. (2016). Research methods in healthcare epidemiology and antimicrobial stewardship—quasi-experimental designs. Infection control & hospital epidemiology, 37(10), 1135-1140.

Most of the benefits are pragmatic: less cost, potential for retrospective analysis, bypassing ethical considerations that present barriers to randomized trials, etc.

An example of the type of quasi-randomized design I've often been involved in myself is when the data have already been collected in normal medical practice. Sometimes these patients are difficult if not impossible to consent: an intensive-care unit population, for example. Those patients are very sick and their imminent needs are live-saving care; there may not be time to obtain consent for a randomized trial. However, if practices and policies change over time or are different in different facilities, you can use a quasi-experimental design to determine whether those changes or differences in standard of care have an important impact or not.

  • 1
    $\begingroup$ Thank you very much Bryan Krause! your response was really helpful. what exceptionally grabbed my attention in your answer was ethical problems of randomization and it intrigued me to do more search about it. So, I found out that one of its ethical problems is that some people expect to be benefited from the intervention, but by doing so they might miss the chance( according to link. Now I'm wondering whether that is the solely ethical consideration for randomization, and it will be nice of you if you reply to this question too. $\endgroup$
    – Ali Sirous
    Commented Apr 17, 2020 at 19:49
  • 2
    $\begingroup$ @AliSirous That's one case - I talked about that a bit in my answer. Let's say there is a change in standard of care, and we assume the new one is better. It may not be ethical to randomize people to the old standard of care. But, maybe the change was made based on limited data. We could design a large study where we compare outcomes before the change to outcomes after: this is a quasi-experiment that doesn't expose anyone to study-associated risk, and can be conducted without even getting patient consent after IRB approval. $\endgroup$
    – Bryan Krause
    Commented Apr 17, 2020 at 20:36
  • $\begingroup$ thank you very much again. Now, I comprehend it substantially better. $\endgroup$
    – Ali Sirous
    Commented Apr 18, 2020 at 5:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.